Published by Instant Press on behalf of Fight For A Living Wage.
Ask for the median, not the average. The median is the paycheck sitting in the exact middle of the line, half of workers above it and half below. The average divides total pay by headcount, so a few very large salaries pull it upward until it describes almost nobody on the payroll. Which number a report picks decides whether a wage conversation starts from what most people earn or from a figure the top of the distribution inflated.
What the two numbers measure
Line up every worker at a company from lowest paid to highest paid. The median is whoever stands in the middle. Add every salary together, divide by the number of workers, and you get the average, also called the mean.
Both are correct arithmetic. They answer different questions. The median tells you what a typical worker earns. The average tells you what everyone would earn if you split the payroll evenly. Pay distributions lean hard to one side, so the two numbers separate, and they separate most where the top earners earn the most.
A worked example
Take a ten person business. Eight employees earn $35,000. One manager earns $90,000. The owner takes $600,000.
That payroll totals $970,000. Divide by ten and the average is $97,000. Now line the ten up and find the middle: the fifth and sixth workers both earn $35,000, so the median is $35,000.
Same ten people, same money, and one number says $97,000 while the other says $35,000. The average answers a question nobody asked. Not one employee at that business earns $97,000, and eight of the ten take home less than 40 percent of it.
Scale that pattern up and you get the United States. The Census Bureau publishes both figures for household income, and the average runs well above the median for the same reason the small business did. The upper tail is long, and long tails move averages.
Where the gap shows up in national data
Median household income runs around $80,000, according to Census Bureau figures for 2023. That is the middle of the distribution, and it is the number to hold onto when you read a claim about affordability.
Set it beside housing. The National Association of Realtors and Census data put the median U.S. home sale price at roughly $400,000 to $420,000 in 2024. A typical home now costs about five times a typical household’s annual income. Run the same comparison for the 1980s and you land closer to three times. Neither side of that ratio needs an average, and reaching for one would hide the squeeze instead of showing it.
Pay floors are simpler. The federal minimum wage has sat at $7.25 an hour since 2009, according to the U.S. Department of Labor. Forty hours a week for 52 weeks at that rate grosses $15,080 before payroll taxes. One rate, one number, no distinction to argue about. Most of the wage distribution gives you no such mercy.
Why the Bureau of Labor Statistics publishes both
The BLS Occupational Employment and Wage Statistics program reports mean and median wages for every occupation it tracks, and the pair beats either one alone. A mean sitting far above the median tells you pay inside that job title spreads wide. A mean sitting close to the median tells you most people in the job earn something near the same amount.
Picture two occupations that report the same median wage. In the first, the mean lands a few hundred dollars higher. In the second, the mean lands a third higher. The second occupation holds a cluster of very high earners, and a job seeker who read only the mean would badly misjudge what the work pays at entry.
Wealth stretches the gap further
The Federal Reserve’s Survey of Consumer Finances measures net worth, and there the two numbers diverge far more than they do for income. Wealth concentrates harder than earnings, so mean net worth runs several times median net worth. Any claim about how much the average American has saved means little until it says which statistic it used.
How this plays out in an argument
Someone arguing that workers are doing fine can reach for an average. Someone arguing the opposite can reach for a median. Both cite real government data, both describe the same country, and neither has to lie.
So read for four things. Which statistic you are looking at, median or mean. Whether the unit is a household or an individual worker, because households pool multiple earners and the two figures swap in for each other badly. Which year, since a 2019 figure and a 2024 figure sit on different price levels. And who sits in the denominator: median wages for full time workers leave out part time workers, who skew low paid, and that exclusion lifts the number without improving anyone’s week.
The practical version
For wages and household budgets, use the median. It shrugs off outliers, and the question you care about is what the middle of the country faces.
For totals, use the average. Mean times headcount gets you the full payroll cost of a workforce or the aggregate size of a market. The median will not.
Fight For A Living Wage, a nonpartisan grassroots 501(c)(3), pulls these affordability gaps into one place next to the primary sources behind them, which shortens the work of checking whether a figure quoted at you is a median, an average, or neither.
The distinction looks academic until you notice how often the choice gets made for you. A headline reports average household income and a reader concludes the typical family is comfortable. A recruiter quotes an average salary and a candidate anchors high on a number three colleagues will never see. Asking which one you are holding takes a sentence, and it changes what the rest of the numbers mean.
